In one line
Pregnancy turns the urinary tract into a slow-draining, sugar-rich, mechanically obstructed system, so bacteria that would have been flushed out instead multiply and ascend. Untreated asymptomatic bacteriuria becomes pyelonephritis in 20 to 35% of women, and treating it drops that to 1 to 4%. That is the whole argument for screening a woman who feels perfectly well.
Where this sits. The renal physiology this chapter assumes, filtration, tubular handling and the changes pregnancy imposes on them, is in the Primary chapter on renal, fluid and electrolyte physiology. The consultant layer is the Finals chapter on renal disease in pregnancy, and the woman who becomes septic from her kidney is maternal sepsis.
Why the pregnant urinary tract loses
Outside pregnancy, the main defence against urinary infection is mechanical. Urine is sterile when it leaves the kidney, it flows one way, and it is voided completely and often. Bacteria that reach the bladder are washed out before they can establish. Everything that defence depends on is degraded in pregnancy.
Flow slows. Progesterone relaxes smooth muscle throughout the urinary tract from early pregnancy, before the uterus is large enough to press on anything. Ureteric peristalsis weakens, the ureters dilate, and urine that used to move in a coordinated wave begins to pool.
The ureters are compressed. From the second trimester the enlarging uterus compresses the ureters at the pelvic brim. The uterus is normally dextrorotated, and the left ureter is partly cushioned by the sigmoid colon, so the compression falls hardest on the right. This produces the physiological hydronephrosis and hydroureter of pregnancy, which is present in most pregnant women, is right-sided in most of them, and is a normal finding rather than a diagnosis.
The bladder changes. Capacity increases and tone falls, so residual volume rises. The bladder is displaced upwards and anteriorly, and the vesicoureteric junction loses some of its oblique angle, which promotes vesicoureteric reflux. Bacteria that reach the bladder now have a road upwards.
The urine becomes better food. The renal threshold for glucose falls, so many normal pregnant women have glycosuria. Filtered amino acids and water-soluble vitamins rise. Urinary pH rises. The urine is a richer growth medium than it was.
Assemble those and the clinical pattern falls out without memorising it. Infection is commoner in pregnancy, it ascends more readily, and when it reaches the kidney it is more often on the right.
The three diagnoses, and why the first one matters most
Asymptomatic bacteriuria is significant bacteriuria, conventionally 10⁵ colony-forming units per millilitre or more on culture, in a woman with no urinary symptoms. It affects roughly 2 to 10% of pregnancies. It is usually acquired before pregnancy, not during it, which is why a single early screen catches most of it.
Acute cystitis is the same bacteriuria with symptoms confined to the lower tract: frequency, urgency, dysuria, suprapubic discomfort, sometimes haematuria. She is systemically well and afebrile.
Acute pyelonephritis is infection of the renal parenchyma: fever of 38.0°C or higher, rigors, loin pain, renal angle tenderness, nausea and vomiting, with or without lower tract symptoms. She looks ill.
The reason asymptomatic bacteriuria earns a screening programme, when almost no other asymptomatic condition does, is the size of the effect. Untreated, 20 to 35% of these women develop pyelonephritis. Treated, 1 to 4% do. Treatment also reduces low birth weight and preterm delivery. Few interventions in antenatal care have that leverage.
What pyelonephritis actually does
It is worth being precise about why a kidney infection is dangerous in pregnancy, because "she might get sepsis" understates it.
Pyelonephritis is the commonest non-obstetric cause of admission and of septic shock in pregnancy. Gram-negative organisms release endotoxin, which produces vasodilatation, capillary leak and myocardial depression. On top of a pregnancy that has already lowered systemic vascular resistance and colloid oncotic pressure, the woman decompensates faster than a non-pregnant patient with the same infection.
Three specific consequences:
- Pulmonary injury. A small but important proportion of women with obstetric pyelonephritis develop acute respiratory distress syndrome, from endotoxin-mediated capillary leak, often made worse by enthusiastic fluid resuscitation. A woman with pyelonephritis who becomes breathless is not developing pneumonia; she is leaking into her lungs. Fluid her carefully and monitor her saturation. See Shock management and Arterial blood gas.
- Preterm labour. Endotoxin and the inflammatory cytokine response stimulate prostaglandin release and uterine activity. Pyelonephritis is a genuine and treatable cause of preterm contractions, and the treatment is the antibiotic, not only the tocolytic. See Preterm birth and pprom.
- Haemolysis and renal dysfunction. Endotoxin-mediated red cell destruction produces a fall in haemoglobin in a substantial minority, which matters in a population already anaemic. Transient reduction in creatinine clearance is common. See Anaemia in pregnancy.
Screening: when, and with what
Screen every woman with a urine culture once, early in antenatal care. This is the ACOG position and the IDSA position, and it is the practice that the effect size above justifies. There is insufficient evidence to recommend routine repeat screening after a negative initial culture, so a single early culture is the standard, with repeat testing driven by symptoms or by a previous positive.
Dipstick is not a substitute. Nitrite depends on the organism being a nitrate-reducer, which excludes Enterococcus and group B Streptococcus, two organisms that matter here. Leucocyte esterase detects pyuria, which asymptomatic bacteriuria often does not produce. The sensitivity of dipstick for asymptomatic bacteriuria in pregnancy is too low to rely on. Where culture is genuinely unavailable, a positive dipstick is worth acting on and a negative one is worth ignoring, which is a poor test in both directions and should be recognised as a resource constraint rather than a clinical standard.